ACTIVE OBJECT DETECTION

G. de Croon

2007

Abstract

We investigate an object-detection method that employs active image scanning. The method extracts a local sample at the current scanning position and maps it to a shifting vector indicating the next scanning position. The method’s goal is to move the scanning position to an object location, skipping regions in the image that are unlikely to contain an object. We apply the active object-detection method (AOD-method) to a face-detection task and compare it with window-sliding object-detection methods, which employ passive scanning. We conclude that the AOD-method performs at par with these methods, while being computationally less expensive. In a conservative estimate the AOD-method extracts 45 times fewer local samples, leading to a 50% reduction of computational effort. This reduction is obtained at the expense of application generality.

Download


Paper Citation


in Harvard Style

de Croon G. (2007). ACTIVE OBJECT DETECTION . In Proceedings of the Second International Conference on Computer Vision Theory and Applications - Volume 2: VISAPP, ISBN 978-972-8865-74-0, pages 97-103. DOI: 10.5220/0002044600970103


in Bibtex Style

@conference{visapp07,
author={G. de Croon},
title={ACTIVE OBJECT DETECTION},
booktitle={Proceedings of the Second International Conference on Computer Vision Theory and Applications - Volume 2: VISAPP,},
year={2007},
pages={97-103},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0002044600970103},
isbn={978-972-8865-74-0},
}


in EndNote Style

TY - CONF
JO - Proceedings of the Second International Conference on Computer Vision Theory and Applications - Volume 2: VISAPP,
TI - ACTIVE OBJECT DETECTION
SN - 978-972-8865-74-0
AU - de Croon G.
PY - 2007
SP - 97
EP - 103
DO - 10.5220/0002044600970103